/compression-faq/part1/section-7.html
RLE, and LZ77, LZ78 and every other patent mentioned in that paper has expired.
- Original Message -
From: robert therriault
To: programm...@jsoftware.com
Cc:
Sent: Thursday, September 18, 2014 11:39 AM
Subject: Re: [Jprogramming] LZ77
Unintended but not
AFAIK whether zlib/deflate itself is patent-free is still
an unknown, because it has not yet trial tested.
patent for LZW expired but other variations of LZ77 might not.
Чт, 18 сен 2014, robert therriault написал(а):
> Hi Bill
>
> Nice joke in your last two sentences.
>
> You wouldn't want to
Unintended but not unrelated. :-)
And now I think that patent law has already sucked up enough space in this
forum on this fine morning, I now return to my regular 'programming'
(intended). :-)
cheers, bob
On Sep 18, 2014, at 8:31 AM, Raul Miller wrote:
> Pun intended?
>
> http://www.amer
Pun intended?
http://www.americanbar.org/aba.html
Thanks,
--
Raul
On Thu, Sep 18, 2014 at 11:22 AM, robert therriault
wrote:
> The best form of self defence in a bar fight is to have left the bar 10
> minutes before the fight starts. I learned that in Tai Chi and I think
> software patent
The best form of self defence in a bar fight is to have left the bar 10 minutes
before the fight starts. I learned that in Tai Chi and I think software patent
law is a lot like a bar fight. ;-)
cheers, bob
On Sep 18, 2014, at 8:11 AM, Raul Miller wrote:
> It's unlikely that most of those pat
It's unlikely that most of those patents have merit.
First, there's prior art (the LZ77 patent, for example - which is now
expired). And second, patents do not cover variations which would have
been obvious to someone with ordinary skill in the art before the
filing date of the patent.
In the con
Hi Bill
Nice joke in your last two sentences.
You wouldn't want to do this on the chance that J ends up with revenue and
could become a target (especially if your joke were produced as evidence).
Legal action ends up being very expensive to defend in time as well as money.
Also, things tend t
Thanks. This is not urgent because zlib.so should usually
available.
Also as stated in rfc, many variation of LZ77 were patented so
it suggested implementors to follow the guidelines (3 bytes at a
time, hash etc) which is patent free. If the improvement in
speed is signaficant then we may ignore
I can try taking a look at the code, one of these days, if you want me to.
(Not right this second though.)
Thanks,
--
Raul
On Thu, Sep 18, 2014 at 5:45 AM, bill lam wrote:
> Thanks Raul,
>
> I followed you suggestion to use 2 generations of hash vector,
> Its speed is faster now although sti
Thanks Raul,
I followed you suggestion to use 2 generations of hash vector,
Its speed is faster now although still far behind zlib.so.
Ср, 17 сен 2014, Raul Miller написал(а):
> I would ignore the stuff about "hashes" and "hash chains" and "singly
> linked". Hashes are nice sometimes but other ti
I would ignore the stuff about "hashes" and "hash chains" and "singly
linked". Hashes are nice sometimes but other times they're just a
disaster. Sometimes that's better than the alternative, but their
marketing is sometimes significantly better than their actual
behavior.
Anyways, what's really h
The LZ77 variation used in deflate is at the bottom (copied from rfc
1951). I tried to translate it into J using a hash function which
interprets XYZ as the lower 3 bytes of a 4 byte integer. My J
implementation worked but efficient is rather poor. I think the
inefficiency of my implementation c
12 matches
Mail list logo
